skills.rest
zlovexforever
zlovexforeverCOMMUNITY
Explore all skills by this author →
GitHub

skill-creator

Generate SKILL.md content and run trigger evals with pass-rate benchmarks.

Updated Mar 16, 2026
Search Tags:#skill creation#benchmarking#evaluation#iterative improvement#trigger optimization#Claude Code skills#markdown instructions
One-click install
npx skills add https://github.com/zlovexforever/zlx-claude-skills --skill skill-creator-zlovexforever
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/zlovexforever/zlx-claude-skills/tree/main/skill-creator
Command: npx skills add https://github.com/zlovexforever/zlx-claude-skills --skill skill-creator-zlovexforever

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and assets (resource) and references (resource) components.

What problem does it solve?

This Skill helps you turn an idea into a working skill, then iteratively improve that skill using real triggering tests and measured performance.

Core Features & Use Cases

  • Skill drafting & iteration: Convert user intent into a new or improved SKILL.md, plus a structured loop for refinement.
  • Eval-driven validation: Generate test prompts, run them, and use quantitative pass/fail metrics to guide changes.
  • Benchmarking and analysis: Aggregate results (including timing and tokens) and surface patterns that averages can hide.
  • Description optimization: Produce a higher-triggering description by testing many candidate trigger queries.

Quick Start

Use the skill when you have a draft SKILL.md (or want to create one) and want to evaluate whether Claude triggers it correctly, then improve its description and overall performance.

Frequently Asked Questions about skill-creator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I optimize Claude Code skills to trigger correctly and avoid false triggers?▼

Optimize Claude Code skills by designing markdown instructions, generating realistic trigger evals, and iteratively refining the description to increase correct triggering while minimizing false positives.

What is the best way to benchmark Claude skill performance using quantitative metrics?▼

Benchmark Claude skill performance by running an eval loop that collects outputs, grades expectations with evidence, and aggregates statistics like pass rate, time, and tokens to surface hidden patterns.

How do I create a new skill from scratch and validate its performance?▼

Create a new skill from scratch by converting user intent into a structured markdown file, generating realistic test prompts, and using pass or fail metrics to guide iterative refinement.

Can I use description optimization to improve skill triggering accuracy with held-out test queries?▼

Improve skill triggering accuracy by running a description optimization loop that tests many candidate trigger queries against held-out test queries to produce a higher-triggering description.

Does this skill creation workflow require existing markdown instructions to start?▼

The skill creation workflow does not require existing markdown instructions, allowing you to draft new skills from user intent or modify existing markdown files to improve overall performance.

Why does my Claude skill fail to trigger accurately in real-world scenarios?▼

Claude skills fail to trigger accurately due to poorly structured descriptions, which can be fixed by generating realistic trigger evals, running comparisons, and iteratively refining instructions.

Related Skills

Explore More Product & Management Skills→
hiddink-aihiddink-ai
1

hiddink-harness-takeover

Parse YAML frontmatter and markdown to generate structured .spec.md files.

Official
Intermediate
JonathanMitchell1234JonathanMitchell1234

skill-creator

Create and refine AI skills with SKILL.md files and test cases.

Community
Advanced
beyonaibeyonai
45

skill-optimizer

Optimize agent skill instructions through a validation-gated feedback loop.

Community
Advanced
teamniteoteamniteo
5

skill-creator

Develop, evaluate, and refine custom AI skills with iterative testing and benchmarking.

Official
Advanced